Your web browser does not support JavaScript, but it does not affect browsing through the rest of the web site.

ITRI Welcomes New AI Lifestyle at COMPUTEX 2018 and Exhibits 12 Innovations

Font size:
Small
Medium
Large

ITRI Welcomes New AI Lifestyle at COMPUTEX 2018 and Exhibits 12 Innovations

Can you imagine having a robot at home with an intelligent vision system that can play Scrabble with you? Or what about embarking on a new trend by experiencing a computer vision-based self-checkout system that relies solely on a scanner? ITRI this year at its Smart Systems Pavilion in the SmarTEX exhibition area is showcasing 12 technology innovations under the theme "Welcoming the New AI Lifestyle." These include Intelligent Vision System for Companion Robots, Computer Vision Based Self-Checkout System, Micro LED, ITRI MusicAI, Hybrid Power Drone with High Payload and Duration, and Long-acting Drone and Aerial Photography Intelligence Analysis Technology. All of these technologies and applications are expected to inject new momentum into industrial innovation.

 

The emphasis of this year's COMPUTEX is on AI-related technology. ITRI exhibits its achievements in the integration of optoelectronics, ICT, and machinery in the form of smart machinery, lifestyle, and entertainment solutions. These reflect ITRI's efforts in working in tandem with the government's 5+2 Industrial Innovation Plan, especially the achievements in promoting smart machinery and Asian Silicon Valley.

 

ITRI President Edwin Liu remarked that Taiwan is taking advantage of the arrival of the AIoT era by focusing on the manufacturing sector's AI strengths in endpoint data, various databases, and semiconductor core computing technologies. Taiwan should be able to be a key partner in vertical integration or ecosystem leaders, utilizing smart systems and services to create new types of technology service models. ITRI, he said, will integrate AI, big data, and hardware & software technology to form an innovation-oriented platform that can be utilized in a flexible manner. Based on this open and innovative service platform, various smart systems are being developed to assist in industrial transformation.   

Intelligent Vision System for Companion Robots integrates user needs for cross-disciplinary innovations

ITRI's Industrial Economics and Knowledge Center (IEK) forecasts the global market for intelligent robots will grow to US$33.6 billion by 2021. In light of anticipated manpower shortages, sales growth for service-oriented robots needed in smart homes and healthcare will be even faster. ITRI’s Intelligent Vision System for Companion Robots enables a robot to adapt to changing conditions and act accordingly. Not only can it distinguish between various Scrabble cubes, words, and their locations, but also recognize individual letters and the words spelled by other players. The same intelligence applies in various industrial-automation and Industry 4.0 applications; for example, a robot equipped with ITRI’s Intelligent Vision System could differentiate objects based on their size, positions, and surface details even if their dimensional information is similar. In addition to perceiving different objects based on size, shape, color and location, it also enables a robot to grip, move and interact with the objects, while avoiding collisions with other objects.

 

Twitch celebrity Lilballz commented that the experience playing Scrabble with ITRI’s Intelligent Vision System for Companion Robots is jaw-dropping. Unlike eSports matches, the robot is able to interact in person with its incredibly quick decision-making and movements, which also depicts ITRI’s strong suits in AI.

ITRI’s Intelligent Vision System for Companion Robots is back in action at Computex 2018.
ITRI’s Intelligent Vision System for Companion Robots is back in action at Computex 2018.

Computer Vision Self-Checkout System ushers in age of smart retail sales

ITRI's Computer Vision Based Self-Checkout System is an innovation reflecting the government's digital economy initiative, which is part of its broader industrial innovation policy. Through the use of deep learning technology to detect the location of the product, and in tandem with the shape, color and writing of the item, the system can recognize the type and quantity of the product purchased. It then utilizes computer vision technology to replace the need for a human to scan the bar code at checkout. Customers only need to place the item on the computer vision checkout counter and the machine can automatically recognize the product, quantity and price, all of which are displayed on a screen. The entire process requires only five seconds. In the future, this system can be attached to a point of sale, providing a lower cost hardware investment compared with existing systems overseas, thus creating a higher level of acceptance in the marketplace. Besides utilizing computer vision technology as its core, the system also uses deep learning models to provide training in production identification, and at present achieves 97% accuracy in this regard. The product development team overcame recognition challenges caused by different lighting sources or cans or boxed drinks that were stood upright. Different algorithms and alert signals helped to tackle these issues.

Computer Vision Based Self-Checkout System integrated deep learning models, opening up endless opportunities for future applications.
Computer Vision Based Self-Checkout System integrated deep learning models, opening up endless opportunities for future applications.